Latest data from the Evolut Low Risk trial, comparing Medtronic’s Evolut supra-annular, self-expanding, transcatheter aortic valve implantation (TAVI) system to surgical aortic valve replacement (SAVR) in patients at low surgical risk, have shown that, beyond six years, patients in the TAVI arm of the trial required reintervention more often than those undergoing SAVR, driven by an increased incidence of aortic regurgitation.
The trial’s six-year outcomes are detailed in the Journal of the American College of Cardiology (JACC) in a paper by John Forrest (Yale School of Medicine, New Haven, USA) and colleagues on behalf of the Evolut Low Risk investigators.
Evolut Low Risk is a randomised, multicentre trial assessing the safety and efficacy of the Evolut TAVI system versus surgery in patients who had a predicted 30-day mortality risk <3%, as assessed by a heart team—one of the largest clinical trials to investigate TAVI outcomes among patients deemed to be at low surgical risk. A total of 1,414 patients were randomised, with 730 receiving TAVI with either a Medtronic Evolut R, PRO, or CoreValve and 684 undergoing surgery.
Out to five years, the trial has reported comparable rates of all-cause mortality and stroke—the study’s composite primary endpoint—a trend that was maintained at the six-year timepoint, where the primary endpoint rate stood at 23.3% among patients who underwent TAVI and 20.4% among those who underwent surgery, indicating no significant difference between the two approaches.
However, investigators observed an increase in the reintervention rate seen in the TAVI arm at six years, where it stood at 5.5%, compared to 3.3% for surgery. Supplementary data included within the six-year data publication show that up to five years, reintervention rates had been broadly comparable between the two approaches, totalling 1% for TAVI vs. 0.9% for SAVR at three years, 1.3% vs. 1.7% at four years, and 3.3% vs. 2.5% at five years.
The investigators opted to undertake an exploratory, post-hoc analysis to characterise the incidence, aetiology and association of reintervention with mortality using all of the available data through seven years, finding that the reintervention rate for TAVI increased to 9.8% and 6% for surgery.
“This post-hoc reintervention analysis showed comparable rates for reintervention caused by valve stenosis between TAVI and surgery, while reinterventions caused by regurgitation were more frequent in the TAVI group,” Forrest et al report.
Predictor analyses found that younger age was associated with higher reintervention rates for both approaches, whilst TAVI and valve model were also identified as predictive factors for reintervention.
Investigators note that while a substantial portion of the regurgitation-related reintervention signal was seen among patients receiving the Evolut R valve, particularly those receiving the 34mm valve, the totality of events was not limited to this group.
“Reinterventions were observed across valve sizes and platforms, and the smaller Evolut PRO (23–29mm) cohort demonstrated a greater proportion of stenosis-related events. Given sample size, valve availability, and confounding, these observations of valve-specific differences should be interpreted as descriptive rather than mechanistic,” they report.
No formal guidance on maximal balloon sizing existed during the enrolment phase of the trial, but subsequent experience had led to concerns that excessive balloon oversizing may have been associated with structural leaflet injury and subsequent aortic regurgitation. This led to a change in the instructions for use (IFU) for the valve in 2020, cautioning against using a balloon diameter that exceeded the valve waist.
Bench testing performed around this time had demonstrated that off-guidance postdilation could result in injury to the prosthetic valve leaflets at the margin of attachment where they are sutured to the skirt and frame, with the degree of injury potentially varying based on the amount of overexpansion, possibly resulting in immediate or delayed clinical presentation.
“Our post-hoc analysis suggested that among TAVI-treated patients, off-guidance postdilation was associated with a higher incidence of reintervention for aortic regurgitation compared with TAVI valves with on-guidance or no postdilation; however, these analyses are observational and hypothesis-generating,” the study’s authors write. “Additionally, our data suggest that postdilation per se is not harmful and it should be noted that postdilation with an appropriate balloon (on-guidance) results in reintervention rates that are similar to TAVI with no postdilation.”
Although off-guidance postdilation may contribute to leaflet injury, regurgitation-related reintervention can still occur in patients without off-guidance postdilation, they noted, adding that in the analysis, the majority of reinterventions occurred in patients who did not undergo off-guidance postdilation.
“This analysis highlights the critical importance of transparent, robust, long-term clinical and real-world data in low-surgical-risk patients undergoing TAVI,” Forrest et al write. “The Evolut Low Risk trial is one of the largest and longest-running trials for TAVI in low-risk patients; yearly follow-up and data review will be critical for monitoring the composite endpoint, mortality rates, structural valve deterioration, and the incidence of reintervention with TAVI compared with surgery in this patient population.”
In an editorial article published alongside the results in JACC, Ziad Ali (St Francis Hospital, Roslyn, USA) cautions against an “overly narrow” interpretation of the data as being suggestive of the higher reintervention rates being tied to certain valve types or sizes.
“Meaningful event rates are observed across valve sizes, across postdilation strata, and among recipients of other valve generations, including Evolut PRO, where the pattern of failure appears more frequently stenosis-related,” he writes. “These observations highlight the heterogeneity of late valve failure phenotypes and underscore the complexity of interpreting device performance across evolving platforms, sizes, and implantation techniques.”
On the association between off-guidance postdilation and later regurgitation-related reintervention, he writes that the emerging data regarding leaflet stress with balloon oversizing add to this “intriguing” observation, but notes that as the decision to postdilate is typically taken intraprocedurally, the analyses are vulnerable to confounding by indication and reverse causality.
“For interventionalists, the message is not that avoiding postdilation will eliminate risk. Nor is it that one valve size or generation is solely responsible,” Ali writes. “The more important lesson is that long-term valve durability is nuanced, influenced by anatomy, device design, implantation technique, and perhaps factors we do not yet fully understand. The dataset does not allow us to declare a single cause. It does, however, compel us to look carefully.”
In a further editorial, Aakriti Gupta (Cedars-Sinai Medical Center, Los Angeles, USA) and David J Cohen (St Francis Hospital and Heart Center, Roslyn, USA) write that the type of reintervention needed may be as important as the incidence of reinterventions overall, at least from the patients’ perspective.
“In the Evolut Low Risk trial, reintervention after initial TAVI often meant explant surgery (65%)—the very procedure many patients believed they were going to avoid. Although 30-day mortality after reintervention was infrequent in the study population (3.3%), it is important to remember that these procedures were performed in a low-risk population and at experienced centres,” they write. Though this figure stands lower than reported rates of explant-related mortality seen elsewhere, it underscores that valve durability is likely to be an important consideration when subsequent reintervention is likely to entail a major operation, they note.
The study offers important insights into the mechanism of valve failure, Gupta and Cohen state, with rates of reintervention for stenosis similar between the two arms, but regurgitation-related reintervention more frequent after TAVI. “This asymmetry is clinically important, since stenosis typically evolves gradually and can be tracked by monitoring gradients on surveillance imaging, whereas clinically significant transvalvular regurgitation may occur more suddenly and with limited warning,” they add.
Taking the findings in the whole, Gupta and Cohen state that the latest data should not dissuade “thoughtful use of TAVI in low-risk patients”. “Rather, they [the findings] reinforce the critical role of heart-team-based discussions that transparently address both known benefits and remaining uncertainties for patients considering an intervention for treatment of aortic stenosis.”
